Analysis of spatial patterns of public attention on housing prices in Chinese cities: A web search engine approach

Housing price has become one of the most pressing issues facing urban residents in China in recent years and received considerable attention. However, detailed housing price data are often ill-documented or unavailable for the public, thus posing a grand challenge for the study of housing prices in China. Because individuals' Internet search activities can be recorded by web search engines, the analysis of these web search activities in cyber-space may provide a means of better understanding public attention and associated concerns in real geographic space. In this study, we focus on exploring the spatial patterns of public attention on housing price through the analysis of web query activities based on Baidu Index, a Chinese keyword analysis tool from Baidu web search engine. We propose a new index based on keyword query outcome from Baidu search database to analyze spatially heterogeneous patterns of housing price attention from 19 large and medium-sized cities in China. We evaluate the spatial network structure of housing price attention, and develop a new index to measure the intensity of interaction relationships among cities of interest. Our results show that spatial interactions of housing price attention between cities evaluated using the new method are consistent with those from a gravity model. Meanwhile, as revealed from Baidu Index-based indicators, strong spatial association patterns exist among cities that form urban agglomerations. Further, our results demonstrate that the web search engine approach, based on the coupling of cyber-space and geographic space, provides solid support for the study of housing price attention and its spatially explicit patterns in China. (C) 2016 Elsevier Ltd. All rights reserved.